Sounds such as melody, sound effects, etc. are generally defined by a signal having regular waveforms such as sinusoidal waveform, triangle waveform, rectangle waveform, etc. Sounds are conventionally converted from analog signals to digital signals of PCM data to be stored in a melody IC, wherein a repetitive PCM method is utilized to decrease an amount of the stored data.
The repetitive PCM method is a method, in which PCM data of segmental waveforms and repetition numbers of the segmental waveforms are stored, when the same or similar waveforms continue repetitively in sections of a sound signal each including plural periods.
Conventionally, a personal computer is used for a system for developing sounds. Practically, sound signal waveform information such as shapes of segmental waveforms, repetition numbers of the segmental waveforms, amplitudes of the segmental waveforms, and frequencies of the segmental waveforms is supplied to the personal computer, in which the information is edited, so that a waveforms of the sound signal are obtained to be displayed on a CRT of the personal computer. A desired waveform of a sound signal thus obtained is supplied to be stored in a melody IC, etc.
However, the conventional system for developing sounds has a disadvantage in that information supplied to the personal computer is difficult for an operator to be analyzed, so that it takes a long time in developing a desired sound. Even if a spectrum analyzer is used to analyze a waveform of a predetermined sound signal, the disadvantage is not overcome, because the spectrum analyzer only provides limited information. Otherwise, when information is supplied to the personal computer in try and error method without using the spectrum analyzer, the above described difficulty will be increased.